Wind speed trends

Figure 3-4 shows the direction-wise wind speed trends over Toronto at the σ995 level and

the quantitative details of the observed trends are given in Table 3-2. Only winds coming in from the 90°, 150°, 180°, and 330° directions had negative trends. However, these trends are not statistically significant, as indicated by the zero values of H and near-zero Z

statistics. The strongest positive trends with Sen’s slope equal to 0.021 m s-1 _year-1 _and

0.015 m s-1 _year-1 _{are detected for the 𝑉}

𝜎99530° and 𝑉𝜎99560° winds, respectively. Therefore, the

𝑉_𝜎99530° _{winds have increased their speed for 1.41 m s}-1 _{in the period from 1948 to 2014 (0.21}

m s-1 _{per decade).The windiest directions, 𝑉}

𝜎995240° and 𝑉𝜎995270°, have also experienced

statistically significant upward trends. Sen’s slope of the trend for these two winds is the same and equal to 0.008 m s-1 _year-1_{. Therefore, the 𝑉}

𝜎995240° winds have increased for 0.54 m

s-1 in the analyzed 67-year period. It is important to note that the positive trends of the

𝑉_𝜎995240° _{and 𝑉}

𝜎995270° winds are significant at the 99% confidence level, as indicated by H=3

values in Table 3-2. The positive trend of northern winds is also fairly significant (H=1). The moving averages in Figure 3-4, however, seem to indicate that the windiest directions (210°, 240° and 270°) were fairly trendless after about year 2000. The strong upward trends for 𝑉_𝜎9950° , 𝑉_𝜎99530° and 𝑉_𝜎99560° are also noticeable in the moving average trend.

122 Figure 3-4. Mean annual wind speeds for each wind direction. The blue lines are Sen’s

slope (m s-1 _year-1_{) and the red dashed lines are trends at the 95% confidence intervals}

and the green lines are a 5-year moving average. See Table 3-2 for additional details.

The trend analysis of the omnidirectional mean annual wind speed series is presented in Figure 3-5 and in the last row in Table 3-2. A positive and statistically significant trend at the 95% confidence level (H = 2) is observed. Sen’s slope is equal to 0.003 m s-1 year-1. This trend reflects as a wind speed increase of 0.2 m s-1 in the period 1948-2014. The most important contributors to this small wind speed increase are positive trends of the 𝑉_𝜎99530° ,

𝑉_𝜎99560° _{, 𝑉}

𝜎995240° and 𝑉𝜎995270° winds, as described in the previous paragraph. Note, however, that

123 (Qmin99 = 0). The 5-year moving average series in Figure 3-5 has an increasing trend until

the 1990s and is relatively trendless afterwards. A periodicity in the moving average line can be observed. The first six peaks (in years 1952, 1961, 1969, 1977, 1985 and 1991) have the period between 7 and 9 years, whereas the periodicity of the last three peaks is 11 years (years 1991, 2003 and 2014). This result shows the presence of the interdecadal variability of the mean annual wind speeds above Toronto. Figure 3-4 further depicts that the strongest winds above Toronto were present in the early 1990s and the late 1980s.

Table 3-2. Trend analysis of the mean seasonal wind speeds and the mean annual wind speeds for each wind direction and overall (last row). Subscripts denote different significance level, B is the offset of the linear trend line (m s-1), Q is the slope of the trend (m s-1 year-1) and the meaning of other symbols is given in the text. All values are given at the σ₉₉₅ level. Graphical representations of these trends are shown in Figure 3-4, Figure 3-5 and Figure 3-6.

Wind speed

series Z H Q Qmin99 Qmax99 Qmin95 Qmax95 B Bmin99 Bmax99 Bmin95 Bmax95

0° 1.83 1 0.007 -0.003 0.016 -0.001 0.013 3.54 3.91 3.27 3.85 3.30 30° 5.26 4 0.021 0.011 0.031 0.014 0.028 3.28 3.66 2.91 3.57 3.02 60° 4.05 4 0.015 0.006 0.025 0.007 0.022 3.32 3.59 3.05 3.53 3.14 90° -0.34 0 -0.001 -0.011 0.010 -0.008 0.007 3.67 4.01 3.34 3.86 3.43 120° 0.32 0 0.001 -0.007 0.010 -0.005 0.008 3.53 3.83 3.31 3.75 3.37 150° -0.23 0 -0.001 -0.010 0.008 -0.008 0.006 4.24 4.53 3.88 4.44 3.97 180° -0.80 0 -0.003 -0.011 0.005 -0.009 0.003 4.88 5.15 4.69 5.11 4.73 210° 1.50 0 0.004 -0.003 0.010 -0.001 0.009 5.11 5.36 4.95 5.28 4.98 240° 2.74 3 0.008 0.001 0.017 0.003 0.015 5.37 5.59 5.12 5.55 5.18 270° 2.59 3 0.008 0.000 0.015 0.002 0.014 5.08 5.37 4.87 5.31 4.89 300° 0.36 0 0.001 -0.005 0.007 -0.004 0.005 4.74 4.97 4.60 4.91 4.65 330° -0.14 0 -0.001 -0.009 0.008 -0.006 0.006 4.29 4.53 4.03 4.46 4.10 Spring -0.95 0 -0.002 -0.006 0.003 -0.005 0.002 4.77 4.89 4.68 4.87 4.71 Summer -0.52 0 -0.001 -0.005 0.004 -0.004 0.003 3.91 4.09 3.77 4.05 3.79 Fall 4.21 4 0.012 0.006 0.019 0.007 0.017 4.65 4.89 4.46 4.81 4.51 Winter 1.67 1 0.004 -0.002 0.011 -0.001 0.009 5.46 5.66 5.21 5.61 5.28 Annual 2.45 2 0.003 0.000 0.007 0.001 0.006 4.71 4.86 4.59 4.83 4.61

124 Figure 3-5. Mean annual wind speeds for all wind directions and the associated trends.

See Figure 3-4 and Table 3-2 for nomenclature and further details.

Further analysis is conducted in order to determine the seasonal wind speed trends. The results presented in Figure 3-6 and Table 3-2 highlight some interesting findings. Namely, wind speeds in the warm half of the year (spring and summer) have almost negligible and statistically not significant trends with slightly negative Sen’s slopes. On the other hand, wind speeds in the cold part of the year (fall and winter) have experienced a strong and positive trend of 0.08 m s-1 per decade on average (0.54 m s-1 over the whole analyzed period). The positive trend in fall season is significant at the 99.9% confidence level (H = 4) while the winter wind speed trend is significant with the 95% statistical confidence. It can also be seen that the trend of mean annual wind speed is similar to the trend of winter winds. Furthermore, trend in the annual winds is half of the average of the winter and fall trends. This result is somewhat expected, since an average of the seasonal trends should

125 approximately, but not necessarily identically, result in the observed trend of the annual wind. The observed positive trends of the winter and fall winds, as well as the positive trends of the windiest directions, are for example important in wind energy. Namely, these positive trends might contribute to long-term stability and sustainability of urban wind energy projects in the Toronto and nearby regions as the future positive trends in surface wind speeds in this region are predicted by climate change models (McInnes et al. 2011). Similar to the moving average trend in Figure 3-4, there seem to be an absence of moving average trends in the summer, fall and winter seasons after approximately the year 2000. On the contrary, a negative trends in the last about 15 years is noticed in the spring season.

Figure 3-6. Mean annual wind speeds per season. See Figure 3-4 and Table 3-2 for nomenclature and further details.

126